Github Repository - https://github.com/Aerishgaba/Data-Analysis

##Loading the dataset

IMDB_Action <- read.csv("/Users/aerishgaba/Desktop/Projects/IMDB_Action.csv")

1. Structure of the Dataset

str(IMDB_Action)

## 'data.frame':    7800 obs. of  11 variables:
##  $ Title               : chr  "Anora" "I'm Still Here" "Flow" "Once Upon a Time... in Hollywood" ...
##  $ Title_URL           : chr  "https://www.imdb.com/title/tt28607951/?ref_=sr_t_1" "https://www.imdb.com/title/tt14961016/?ref_=sr_t_2" "https://www.imdb.com/title/tt4772188/?ref_=sr_t_3" "https://www.imdb.com/title/tt7131622/?ref_=sr_t_4" ...
##  $ Image               : chr  "https://m.media-amazon.com/images/M/MV5BYThiN2M0NTItODRmNC00NDhlLWFiYTgtMWM2YTEyYzI3ZTY1XkEyXkFqcGc@._V1_QL75_U"| __truncated__ "https://m.media-amazon.com/images/M/MV5BM2FjMjBiZjgtZDkyYy00YTRlLTk5N2QtODE2ZWIyYWE0Yzg0XkEyXkFqcGc@._V1_QL75_U"| __truncated__ "https://m.media-amazon.com/images/M/MV5BOTM5ODBlOTAtYjcwZi00YzkzLWIzODEtMTM2MTZlNDFmMWU2XkEyXkFqcGc@._V1_QL75_U"| __truncated__ "https://m.media-amazon.com/images/M/MV5BMzMzNmViNjYtN2ViNi00NDM3LWFlMmItNDYyMGIzY2EzZjE2XkEyXkFqcGc@._V1_QL75_U"| __truncated__ ...
##  $ ipclockupoverlay_URL: chr  "https://www.imdb.com/title/tt28607951/?ref_=sr_i_1" "https://www.imdb.com/title/tt14961016/?ref_=sr_i_2" "https://www.imdb.com/title/tt4772188/?ref_=sr_i_3" "https://www.imdb.com/title/tt7131622/?ref_=sr_i_4" ...
##  $ Release.Year        : chr  "2024" "2024" "2024" "2019" ...
##  $ Duration            : chr  "2h 19m" "2h 17m" "1h 25m" "2h 41m" ...
##  $ Rated               : chr  "R" "PG-13" "PG" "R" ...
##  $ Rating              : num  7.6 8.5 7.9 7.6 8.5 8.2 8.7 9.3 8.5 7.7 ...
##  $ Votes               : int  140 90 54 897 610 149 2 3 1 143 ...
##  $ Popularity          : int  91 85 87 84 79 85 74 82 97 94 ...
##  $ Description         : chr  "A young escort from Brooklyn meets and impulsively marries the son of a Russian oligarch. Once the news reaches"| __truncated__ "A mother is forced to reinvent herself when her family's life is shattered by an act of arbitrary violence duri"| __truncated__ "Cat is a solitary animal, but as its home is devastated by a great flood, he finds refuge on a boat populated b"| __truncated__ "As Hollywood's Golden Age is winding down during the summer of 1969, television actor Rick Dalton and his stunt"| __truncated__ ...

df = IMDB_Action

2. Variables of the dataset

variables = names(df)
variables

##  [1] "Title"                "Title_URL"            "Image"               
##  [4] "ipclockupoverlay_URL" "Release.Year"         "Duration"            
##  [7] "Rated"                "Rating"               "Votes"               
## [10] "Popularity"           "Description"

3. First 16 rows of the dataset

top_15 = head(df, 15)
top_15

4. User-defined function to calculate the missing values

missing_values = function(df) {
  n = nrow(df)
  missing = data.frame(
    variable = character(),
    missing_values = numeric(),
    percentage = numeric()
  )
  for (i in 1:ncol(df)) {
    missing[i, 1] = names(df)[i]
    missing[i, 2] = sum(is.na(df[, i]))
    missing[i, 3] = (missing[i, 2] / n) * 100
  }
  return(missing)
}

# Missing values in the dataset
missing_values(df)

5. Filter rows with Votes more then 100

df_filtered = df[df$Votes > 100, ]
df_filtered

6. Independent and Dependent Variables

independent_var <- df$Title
dependent_var <- df$Rating

# Create new dataframe
new_df <- data.frame(independent_var, dependent_var)
new_df

7. Remove missing values from the dataset except Popularity

df_no_na = df[complete.cases(df[, 9]), ]
df_no_na

8. Identify and remove duplicate rows

df_no_duplicates = df_no_na[!duplicated(df_no_na), ]
df_no_duplicates

9. Reorder by column ‘Rating’ in descending order

df_ordered = df_no_duplicates[order(-df_no_duplicates$Rating), ]
df_ordered

10. Rename ‘Release Year’ column to ‘Released’

df_renamed = rename(df_ordered, Released = `Release.Year`)
df_renamed

11. Add new variables in your data frame by multiplying Votes by 2

df_new_var = df_renamed
df_new_var$Votes_x_2 = df_renamed$Votes * 2
df_new_var

12. Create Training Set

Split data (80% training)

set.seed(123)
train_indices <- sample(1:nrow(df_new_var), 0.8 * nrow(df_new_var))
train_set <- df_new_var[train_indices, ]
train_set

13. Summary of the dataset

summary(df_new_var)

##     Title            Title_URL            Image           ipclockupoverlay_URL
##  Length:7338        Length:7338        Length:7338        Length:7338         
##  Class :character   Class :character   Class :character   Class :character    
##  Mode  :character   Mode  :character   Mode  :character   Mode  :character    
##                                                                               
##                                                                               
##                                                                               
##                                                                               
##    Released           Duration            Rated               Rating     
##  Length:7338        Length:7338        Length:7338        Min.   :1.300  
##  Class :character   Class :character   Class :character   1st Qu.:5.300  
##  Mode  :character   Mode  :character   Mode  :character   Median :6.200  
##                                                           Mean   :6.125  
##                                                           3rd Qu.:7.200  
##                                                           Max.   :9.800  
##                                                                          
##      Votes         Popularity     Description          Votes_x_2     
##  Min.   :  1.0   Min.   :  3.00   Length:7338        Min.   :   2.0  
##  1st Qu.:  4.0   1st Qu.: 42.00   Class :character   1st Qu.:   8.0  
##  Median : 21.0   Median : 57.00   Mode  :character   Median :  42.0  
##  Mean   :127.6   Mean   : 57.65                      Mean   : 255.3  
##  3rd Qu.:129.0   3rd Qu.: 73.00                      3rd Qu.: 258.0  
##  Max.   :997.0   Max.   :100.00                      Max.   :1994.0  
##                  NA's   :3936

14. Mean, Mode and Median of Rating column

mean_rating = mean(df_new_var$Rating, na.rm = TRUE)
mode_rating = as.numeric(names(sort(table(df_new_var$Rating), decreasing = TRUE)[1]))
median_rating = median(df_new_var$Rating, na.rm = TRUE)

mean_rating

## [1] 6.124516

mode_rating

## [1] 6.1

median_rating

## [1] 6.2

15. Scatterplot for Rating and Released column

ggplot(df_new_var, aes(x = Rating, y = Released)) +
  geom_point() +
  labs(title = "Scatterplot for Rating and Released column")

16.Bar plot for Rating column

ggplot(df_new_var, aes(x = Rating)) +
  geom_bar() +
  labs(title = "Bar plot for Rated and Rating column by binning the Rating column")

#Barplot for Rating column
ggplot(df_new_var, aes(x = Rated)) +
  geom_bar(fill = "darkred") +
  labs(title = "Bar plot for Rated column")

Converting Released column to numeric

df_new_var$Released = as.numeric(as.character(df_new_var$Released))
summary(df_new_var)

##     Title            Title_URL            Image           ipclockupoverlay_URL
##  Length:7338        Length:7338        Length:7338        Length:7338         
##  Class :character   Class :character   Class :character   Class :character    
##  Mode  :character   Mode  :character   Mode  :character   Mode  :character    
##                                                                               
##                                                                               
##                                                                               
##                                                                               
##     Released      Duration            Rated               Rating     
##  Min.   :1915   Length:7338        Length:7338        Min.   :1.300  
##  1st Qu.:1991   Class :character   Class :character   1st Qu.:5.300  
##  Median :2009   Mode  :character   Mode  :character   Median :6.200  
##  Mean   :2003                                         Mean   :6.125  
##  3rd Qu.:2018                                         3rd Qu.:7.200  
##  Max.   :2025                                         Max.   :9.800  
##  NA's   :1                                                           
##      Votes         Popularity     Description          Votes_x_2     
##  Min.   :  1.0   Min.   :  3.00   Length:7338        Min.   :   2.0  
##  1st Qu.:  4.0   1st Qu.: 42.00   Class :character   1st Qu.:   8.0  
##  Median : 21.0   Median : 57.00   Mode  :character   Median :  42.0  
##  Mean   :127.6   Mean   : 57.65                      Mean   : 255.3  
##  3rd Qu.:129.0   3rd Qu.: 73.00                      3rd Qu.: 258.0  
##  Max.   :997.0   Max.   :100.00                      Max.   :1994.0  
##                  NA's   :3936

# Removing missing values from Released column
df_new_var = df_new_var[complete.cases(df_new_var$Released), ]
summary(df_new_var)

##     Title            Title_URL            Image           ipclockupoverlay_URL
##  Length:7337        Length:7337        Length:7337        Length:7337         
##  Class :character   Class :character   Class :character   Class :character    
##  Mode  :character   Mode  :character   Mode  :character   Mode  :character    
##                                                                               
##                                                                               
##                                                                               
##                                                                               
##     Released      Duration            Rated               Rating     
##  Min.   :1915   Length:7337        Length:7337        Min.   :1.300  
##  1st Qu.:1991   Class :character   Class :character   1st Qu.:5.300  
##  Median :2009   Mode  :character   Mode  :character   Median :6.200  
##  Mean   :2003                                         Mean   :6.124  
##  3rd Qu.:2018                                         3rd Qu.:7.200  
##  Max.   :2025                                         Max.   :9.800  
##                                                                      
##      Votes         Popularity     Description          Votes_x_2     
##  Min.   :  1.0   Min.   :  3.00   Length:7337        Min.   :   2.0  
##  1st Qu.:  4.0   1st Qu.: 42.00   Class :character   1st Qu.:   8.0  
##  Median : 21.0   Median : 57.00   Mode  :character   Median :  42.0  
##  Mean   :127.6   Mean   : 57.65                      Mean   : 255.3  
##  3rd Qu.:129.0   3rd Qu.: 73.00                      3rd Qu.: 258.0  
##  Max.   :997.0   Max.   :100.00                      Max.   :1994.0  
##                  NA's   :3935

17. Pearson corrrelation between Title and Rating

cor(df_new_var$Released, df_new_var$Rating, method = "pearson")

## [1] -0.1972266

COMP4028 Assignment - Data Analysis

Group 6

2025-03-31

Github Repository - https://github.com/Aerishgaba/Data-Analysis

1. Structure of the Dataset

2. Variables of the dataset

3. First 16 rows of the dataset

4. User-defined function to calculate the missing values

5. Filter rows with Votes more then 100

6. Independent and Dependent Variables

7. Remove missing values from the dataset except Popularity

8. Identify and remove duplicate rows

9. Reorder by column ‘Rating’ in descending order

10. Rename ‘Release Year’ column to ‘Released’

11. Add new variables in your data frame by multiplying Votes by 2

12. Create Training Set

Split data (80% training)

13. Summary of the dataset

14. Mean, Mode and Median of Rating column

15. Scatterplot for Rating and Released column

16.Bar plot for Rating column

Converting Released column to numeric

17. Pearson corrrelation between Title and Rating